期刊名称：Journal of Theoretical and Applied Information Technology
印刷版ISSN：1992-8645
电子版ISSN：1817-3195
出版年度：2018
卷号：96
期号：19
出版社：Journal of Theoretical and Applied
摘要：In the past decade, face recognition is considered as an important biometric type due to its wide applications in practice in terms of authentication. The verification process of a human face is not trivial task especially different face poses are captured to be matched. In this paper, an efficient algorithm for face recognition is proposed. In the beginning, the step is starting by capturing the image of the face, then applying some preprocessing operations, after that feature extraction is applied, which is exploiting Histogram Orientation Gradient (HOG) to build the most representative feature vector for each digital image of the face. Next, the feature vector is passed into binary Support Vector Machine classifier (SVM) to construct a binary-SVM model for one individual in order to either accept or reject this individual. In this research, multiple models of binary-SVM are utilized in this methodology, in which for each individual has its own SVM model, which is deemed as the contribution of this paper. Set of experiments have been conducted to estimate the accuracy and performance of the proposed algorithm by using ORL database, which has 400 images face captured from 40 users each user has 10 different images as variant possess lighting, etc. The result has given accuracy up to 99.23% as successful rate coming from both error types: False Accept Rate (FAR) is 0.25 % and False Reject Rate (FRR) is 0.52 %.
关键词：Face Recognition and verification; Biometrics; Histogram Oriented Gradient (HOG); Support Vector Machine (SVM).
